Year of Award

2025

Document Type

Thesis

Degree Type

Master of Science (MS)

Degree Name

Geosciences

Committee Chair

Joel Harper

Commitee Members

Payton Gardner, Jesse Johnson

Keywords

Firn, Machine Learning, Modeling, Glaciology, Surface Mass Balance, Remote Sensing, L-band Radiometry

Subject Categories

Glaciology

Abstract

The accurate estimation of the amount of meltwater retained by Greenland’s firn layer and its distribution over depth is essential for assessing both the current and future mass balance of the Greenland Ice Sheet. When meltwater infiltrates the firn, it exists as a quantifiable volume of liquid water, measurable as the one-dimensional liquid water amount, hereafter referred to as LWA. A promising new tool for estimating LWA is L-band passive microwave satellite radiometry, which can provide twice-daily estimates using empirical models of firn microwave emission. However, this method faces two primary limitations. First, retrievals are untested against field observations and are uncertain across the range of conditions over which they can accurately estimate LWA. Second, resolving the depth distribution of water requires complex inversion, which can be computationally intensive. Here, we compare L-band–retrieved LWA time series for a low- and high-melt-intensity year to those generated from in situ observations of firn temperatures and densities, the physical model SLF-SNOWPACK initialized with observed firn states, and the regional climate model Modèle Atmosphérique Régional (MAR), to assess the relative performance of L-band retrievals. Our results indicate significant agreement between L-band retrievals and those generated by traditional means, lending confidence to L-band retrievals. Once validated, we train a machine learning model to predict the depth distribution of water from LWA time series, with the final model reaching mean errors of 26% in mass, and 6% in depth of infiltration. Then we demonstrate its application using L-band retrievals finding greater spread in the depth of infiltration of the wet layer in the higher melt year than the lower melt year. We interpret this as prior-year melt altering storage capacity at depth, which can only be accessed through deep infiltration not observed in the lower melt year. This work is divided into three chapters: Chapter 1 contains background information on the physical processes involved and the estimation of LWA; Chapter 2 presents a journal article detailing the assessment of L-band retrievals compared to traditional methods; and Chapter 3 discusses the training of a machine learning model and its application to L- band–retrieved LWA time series.

Download

Included in

Glaciology Commons

Share

COinS
 

© Copyright 2025 Taylor D. Moon